Source: Cordis 在做什么：从 DeepSeek Harness 看 by Antinomie on antinomie.org. The piece is the site’s current homepage post.

DeepSeek Harness (DSH) is an agent harness built on Cordis. The first thing that clicks is not a privileged core loop. Model adapters, tool registries, session logs, and the agent loop itself are all plugins.

This is the first post in my DeepSeek Harness series. I am starting with a close reading of how DSH actually uses Cordis: services, inject, reversible ctx.effect registration, Loader, and the config layers that turn composition into data.

Read More

Most people first meet Claude Code as a chatbot in a terminal. That mental model is too small.

Ask it to explain a bug and it may answer once. Give it a concrete engineering objective and permission to work, and it can read the repository, trace the code, edit files, run tests, inspect the result, and change course when the evidence contradicts its plan. That repeated cycle is what makes the workflow agentic.

This is Part 1 of 6 in my Agentic Coding with Claude Code series. The working model I am keeping is simple: Claude Code is a bounded loop of perceive, reason, act, and feedback, not a longer chatbot reply.

Later posts will cover context, the harness / Agent SDK / stop-reason loop, MCP, evals, and guardrails. This one stays on the loop, the components, and the patterns.

Read More

I learned agentic AI concepts in Python - agent loops, tool calling, multi-agent coordination, production patterns. I even built my own Python assistant - MiuBot - with 10 chat channels, Temporal workflows, and multi-tenant workspaces. Python works well for most of these cases. But while building MiuBot, I kept hitting the same question: what happens when you need to serve many users concurrently over WebSocket with streaming responses? That’s when I discovered GoClaw from the NextLevelBuilder team, and it changed how I think about the problem.

Read More

After months of studying agentic AI patterns in theory - agent loops, tool calling, multi-agent coordination - I wanted to build something real. Not another tutorial project, but an AI assistant I could actually use daily, connected to the chat platforms I already live on. That’s how MiuBot started - forked from Nanobot, then reshaped into something quite different.

Read More

Every company sitting on a data warehouse wants the same thing: let anyone ask questions in plain English and get reliable answers. OpenAI published how they built their internal data agent, and the open-source community responded fast. Here’s a quick summary of three projects pushing this forward.

Read More

This post brings together everything we’ve learned about applying agentic AI to financial services. From role-based prompting through multi-agent coordination, we’ve covered the complete spectrum of techniques for building intelligent financial systems. Here’s the comprehensive overview with production patterns.

Read More

When multiple agents operate simultaneously on trading operations, they must share a consistent understanding of the world state. This post explores the coordination patterns that make multi-agent trading systems reliable: persistent state management, conflict resolution, and multi-agent RAG for comprehensive analysis.

Read More

When building an AI hedge fund, the most critical element is understanding the flow of agent-to-agent communication. Does one agent need to work after another? Can agents work in parallel? What is the specific purpose of each agent? Multi-agent architecture answers these questions by defining specialized agents, shared state communication, orchestration patterns, and routing strategies that together create reliable, intelligent trading systems.

Read More